Ben Bernanke was I think the first to draw attention to the low level of business investment in a speech he gave in 2005. This is more famous for the global savings glut thesis that he put forward to explain the low level of interest rates. This was when "low" interest rates meant a 4% yield on 10-year Treasuries. Of course, the counterpart to high savings and low interest rates, is - or at least should be - high investment rates. But over the last two decades, business investment never reached the levels it did in 2000.
This chart shows U.S business investment as a share of GDP. A low rate of investment is also widespread outside the U.S.
This second chart shows how much capital firms use relative to output. A lower cost of capital should result in a higher ratio. There has been a modest increase since 2000, but this also reflects the fact that post-2000 the profit share in GDP has increased.
To be clear, when I say interest rates are low I don't just mean the yield on government bonds is low; I mean that firms' cost of capital is cheap. The cost of capital is the weighted cost of equity and debt. Back in the late 90's equity valuations were similar to the equity valuations we've seen in the last few years. The biggest difference is the cost of corporate debt. For the safest of borrowers, this has fallen from a 3-4% real interest rate to about zero or even negative real interest rate. Junk bonds have also seen steep declines in real interest rates.
This chart shows the Q ratios - the total value of equities and/or bonds relative to companies' physical assets. Q3 includes the market value of bonds. This measure in 2019 was even higher than in 2000.
One explanation for low business investment is that technology is now more about intangible capital such as software and R&D and less geared towards physical capital like factories and equipment. The problem with this argument is that business investment includes expenditures on software and R&D anyway. Indeed economists actually refer to software and R&D as "physical" capital. They do so because they want to differentiate capital used in the production of goods and services - factories, equipment, software, and R&D - from financial capital: bonds and stocks.
A more sophisticated version of this argument asserts that firms are now investing more in "human" capital instead of physical capital. Human capital is the training and education of workers. Workers are now more educated than they were in 2000, but workers have been getting more educated for decades. As for training, numerous surveys indicate that firms are spending less, not more, on training than they did two decades ago. This shouldn't be a surprise; you'd expect firms to train workers when a factory is built or a production line is upgraded and new equipment is installed. But firms are spending less Capex on equipment and factories than they were in 2000.
The second theory for why business investment has been low is the idea that the American economy had become more monopolized over the last two decades. Since this isn't my theory I'm somewhat having to paraphrase it, but I believe the logic goes as follows: Monopolies, and other firms with market power, find it profit-maximizing to raise prices and sell and produce a smaller quantity of goods. Less production requires less in the way of physical assets and thus less investment. Lower investment, in turn, means less demand for capital and therefore lower interest rates.
On the surface, the market power theory seems it fits the bill. It can explain not just lower investment and interest rates, but the higher U.S corporate profit share as well.
There are big flaws though. To start with, where are all these monopolies? The internet platforms have large market caps, but their U.S revenue and profits are not nearly large enough to account for anything but a minor portion of the aggregate profit increase since 2000. Indeed I can remember back in the '90s when people imagined the internet would be a powerful force for consumers, demolishing business middlemen and just generally flattening the competitive landscape. Even if those promises for the internet were exaggerated one can still believe that the internet helps consumers shop around more easily and thus diminishes firms' market power.
Secondly, the U.S is not alone in having low interest rates and low business investment. Most of the rest of the world does too. But crucially, these other countries do not have an elevated profit share, so the monopoly theory cannot be the explanation for the combination of low business investment and low interest rates in those countries.
Couldn't the monopoly thesis still explain why investment is low in the U.S? After all, something unusual must be going on in the U.S for the corporate profit share to be so high. The decisive evidence against growing market power is the fact that there has been rising excess capacity throughout the U.S economy since 2000. Here is the chart for manufacturing capacity utilization (there are no data series for service sectors). There is a long-term downtrend in utilization, but most countries besides the U.S have flattened out post-2000.
Why would firms, including monopolies, want to expand their production capabilities beyond the level necessary to meet sales? Manufacturing output in 2019 was higher than in 2000, so it is not as if firms raised prices after 2000, and that caused output to fall below 2000 levels of capacity.
What is behind the increase in excess capacity? Ultimately, it's because of the higher corporate profits, but that whole topic deserves its own post.
Alright, you've waited long enough. What is my theory for why firms are not investing more? Because they choose not to. Profitable investment projects exist, but they do not pursue them. Reason? They do not make enough profit to achieve the desired return on capital.
What I have hypothesized is that firms, especially large ones, have come to fetishize achieving a good return on capital. This is the return on capital firms have earned historically, but cannot achieve now because economic growth rates are lower.
What is so important about achieving this return on capital? The key thing to grasp about modern corporations is that the majority of large firms are not owner-managed. The owners are diffuse - investment funds, by and large - who don't have the same insight and tactile knowledge of the firm as the managers of the firm do. This creates a potential problem. How good actually are these managers at running the company?
Looking at what returns on capital these managers are obtaining gives owners a clue as to their effectiveness. If they are lower than competing firms, then there are two possibilities: the managers are being extravagant in their use of capital - they are recklessly expanding into fringe markets, etc, or they are investing the correct amount, and it's their competitors who are underinvesting. Unfortunately, investment funds tend to view the former as more likely.
One can think of the entire universe of large corporations as just one giant firm, where each of these corporations is but a (competing) operating division of that firm. The firm's Board of Directors are the leading investment banks, private equity groups, and activist funds. If the management of one of these operating divisions reports poor returns on capital, questions will be asked about their effectiveness.
This doesn't mean that no one hasn't taken advantage of the low cost of capital to boost investment or to start new projects. Younger, more impulsive, entrepreneurs who are less attached to historical returns on capital targets, are the most likely to adopt capital intensive growth strategies. Over time, as these more entrepreneurial firms grow bigger they could force the incumbents to reply in kind and deploy more capital. Still, so far - and we are over a decade and a half into this process - these firms have not bothered the Firm too much.
My explanation for why firms are not investing more is an incredibly straightforward one. As it turns out I wasn’t the first to suggest that firms would have a minimum return on capital target. Roy Harrod, a post-war British economist, was I think the first economist to speculate that might be so. He gets bonus points for predicting it in advance.
The only other prominent economic figure I’m aware of who shares this perspective is Phillip Lowe, the Governor of Australia’s central bank. In 2019 he gave a speech in which he blamed low business investment on firms having too high hurdle rates. And he didn’t merely observe that fact but also called on Australian firms to reduce those hurdle rates - so far without success.
And that brings me to the subject of what can be done. How can we get firms in the U.S, and elsewhere, to accept lower returns on capital and boost investment? Perhaps all that is needed is for more people to recognize the reason for low investment and firms will spontaneously adjust their hurdle rates. But what if that doesn’t happen? For now, I’ll leave that as an open question.
Gabriel, thanks for your comment. On IRR, there's another reason why firms are reluctant to make use of all cheap capital and that is a lower discount rate implies that they can and should undertake riskier projects - with a higher chance of failure - but no one wants to associate themselves with failure.
My old company had this issue. ROA was so high we couldn’t get them to invest in anything new. But irr for new investments was pretty low. They just ended up consolidating more and more. Think this piece makes strong case for monopoly power though. Remember building a moat is the five key aspects of company, which is like high barrier to entry but also indicative of monopoly pricing. https://t.co/T0H6K98TgJ